headers
Aleksey Gurtovoy | 1 Jan 02:44
Favicon
Gravatar

Re: [mpl][docs]map insert doc: what's t?

Larry Evans wrote:
> On:
> 
>    http://www.boost.org/libs/mpl/doc/refmanual/map.html
> 
> there's the following row in the "Expression Semantics" table:
> 
>    insert<m,x>::type 	A new map equivalent to m except that
> 
>                          at< t, key_type<m,x>::type >::type
> 
>                          is identical to value_type<m,x>::type.
> 
> 
> which is confusing because the 't' in at< t, key_type<m,x>::type >::type
> doesn't occur in the 1st column of that row.  Should it read:
> 
> 
>    insert<m,x>::type 	A new map, t, equivalent to m except that
> 
>                          at< t, key_type<x>::type >::type
> 
>                          is identical to value_type<x>::type.
> 
> ? 

Yep, that's the intended interpretation. Corrected in the sources 
(http://svn.boost.org/trac/boost/changeset/42396).

Thanks!
(Continue reading)

Permalink | Reply | 0 comments |
headers
Andreas Huber | 1 Jan 22:38
Picon
Favicon

[auto_link][thread] Including boost/thread/thread.hpp requires linking against date_time?

Hi

In a Visual Studio 2008 project, when I include the trunk version of 
boost/thread/thread.hpp, it seems I must also link against the date_time 
library. When I build the same program with bjam, while not linking against 
date_time,  the resulting program runs just fine.

So my question is: Why does including thread.hpp require linking against the 
date_time library? It seems that it is not needed at all?

Thanks & Regards,

--

-- 
Andreas Huber

When replying by private email, please remove the words spam and trap
from the address shown in the header. 

_______________________________________________
Unsubscribe & other changes: http://lists.boost.org/mailman/listinfo.cgi/boost
Permalink | Reply | 2 comments |
headers
Johan Råde | 2 Jan 11:58
Picon
Picon

[Review Request: Floating Point Utilities]

I wish to withdraw my review request for the floating point utilities library.
Please remove the library from the review schedule.
The reasons are that:

1. I do not have time to get involved with Boost.
(I have started my own software company.)

2. There does not seem to be that much interest in the library.

I will leave the library in the vault, under the Boost software license.

--Johan Råde

_______________________________________________
Unsubscribe & other changes: http://lists.boost.org/mailman/listinfo.cgi/boost
Permalink | Reply | 8 comments |
headers
Jamie Allsop | 2 Jan 13:05
Picon
Favicon

Re: [Review Request: Floating Point Utilities]

Johan Råde wrote:
> I wish to withdraw my review request for the floating point utilities library.
> Please remove the library from the review schedule.

Please reconsider leaving it in the queue. I'm sure it would be possible 
to find a mutually agreeable time slot for the review since I know time 
is an issue here. I think your library is an important one and would 
like to see it make its way into boost.

> The reasons are that:
> 
> 1. I do not have time to get involved with Boost.
> (I have started my own software company.)

Yes time is often very hard to find :(

> 
> 2. There does not seem to be that much interest in the library.

On the contrary I think there is quite a lot of interest. However it 
requires a somewhat specialist knowledge base to fully appreciate the 
issues it addresses and so many (myself included) probably did not feel 
qualified to offer constructive comments during the earlier discussions. 
I do recall following the discussions however as I did appreciate the 
value of the library.

> 
> I will leave the library in the vault, under the Boost software license.

Thanks for taking the time to write this library in the first place. I
(Continue reading)

Permalink | Reply | 0 comments |
headers
Thorsten Ottosen | 2 Jan 13:19

Re: [ptr_container] How to shift part of a ptr_vector ?

Phil Endecott skrev:
> Herve Bronnimann wrote:
>> Phil:  Pardon my ignorance about ptr_vector, but you may be thinking  
>> about the wrong STL algorithm: try rotate instead of copy:
> 
> Yes, rotate would do something similar to the for loop in my example.
> 
> The point is that ptr_containers can't use mutating std::algorithms; 
> instead, they provide their own implementations of some of them as 
> members.  rotate isn't one of them, so I need some other way of doing it.

You can access mutating iterators over the original containers by 
calling .base() no the iterators.

They are hen iterators over void*&, so you need to cast or use e.g. 
void_ptr_indirect_fun:

http://www.boost.org/libs/ptr_container/doc/indirect_fun.html

-Thorsten
_______________________________________________
Unsubscribe & other changes: http://lists.boost.org/mailman/listinfo.cgi/boost
Permalink | Reply | 2 comments |
headers
Thorsten Ottosen | 2 Jan 13:21

Re: [ptr_container] [serialization] Header dependency between ptr_vector and serialization

Phil Endecott skrev:
> Dear All,
> 
> It seems that the ptr_vector headers include some serialization 
> headers.  I've discovered this because on my Debian system I have 
> installed their libboost-dev package, which includes the header-only 
> libraries such as ptr_container, but not libboost-serialization-dev.  
> Alexey Khudyakov has already filed a Debian bug report:  http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=457654
> 
> So, what should be going on here?  My guess is that a subset of the 
> serialization headers should be included in Debian's header-only 
> libboost-dev package, so that ptr_vector can compile.  I'm not sure if 
> they get any hints for their partitioning from the Boost build system; 
> if they do, then it should be fixed there.  Any thoughts?

1.35 has a non-intrisive implementation of serialization s.t. the two 
libraries are decoupled.

-Thorsten
_______________________________________________
Unsubscribe & other changes: http://lists.boost.org/mailman/listinfo.cgi/boost
Permalink | Reply | 0 comments |
headers
Neil Groves | 2 Jan 14:39

Re: [Review Request: Floating Point Utilities]

Please reconsider leaving it in the queue. I, for one, desperately need
something like this. I would be willing to help with development,
documentation, testing.

The C++ community, in my opinion, needs this library. However I also
understand the difficulty in finding spare time. I certainly would not like
to see this library die. If you have difficulty continuing perhaps we could
arrange a hand over to other developers?

Thanks for the work you have published in the vault.

Regards,
Neil Groves

On Jan 2, 2008 10:58 AM, Johan Råde <rade <at> maths.lth.se> wrote:

> I wish to withdraw my review request for the floating point utilities
> library.
> Please remove the library from the review schedule.
> The reasons are that:
>
> 1. I do not have time to get involved with Boost.
> (I have started my own software company.)
>
> 2. There does not seem to be that much interest in the library.
>
> I will leave the library in the vault, under the Boost software license.
>
> --Johan Råde
>
(Continue reading)

Permalink | Reply | 0 comments |
headers
Stephan Tolksdorf | 2 Jan 15:39
Picon
Picon

Re: [Review Request: Floating Point Utilities]

Johan Råde wrote:
> I wish to withdraw my review request for the floating point utilities library.

Please reconsider. As Neil said, the C++ community really needs this library.

The library is rather small, maybe the Boost.Math maintainers could take over 
the responsibility for it?

Best regards,
   Stephan
_______________________________________________
Unsubscribe & other changes: http://lists.boost.org/mailman/listinfo.cgi/boost
Permalink | Reply | 0 comments |
headers
Paul A Bristow | 2 Jan 15:48

Re: [Review Request: Floating Point Utilities]

>-----Original Message-----
>From: boost-bounces <at> lists.boost.org 
>[mailto:boost-bounces <at> lists.boost.org] On Behalf Of Johan Råde
>Sent: 02 January 2008 10:59
>To: boost <at> lists.boost.org
>Subject: [boost] [Review Request: Floating Point Utilities]
>
>I wish to withdraw my review request for the floating point 
>utilities library.
>Please remove the library from the review schedule.

No! - please leave it there even if you can't give any (or much) time for it.

>The reasons are that:
>
>1. I do not have time to get involved with Boost.
>(I have started my own software company.)

Good luck!

>2. There does not seem to be that much interest in the library.

There may not *seem* to be much but all people who use floating-point (who doesn't?) will *really need it* -
but most of them just
don't know that they do ;-)

>I will leave the library in the vault, under the Boost 
>software license.

Thanks for all your work so far.
(Continue reading)

Permalink | Reply | 3 comments |
headers
Johan Råde | 2 Jan 16:15
Picon
Picon

Re: [Review Request: Floating Point Utilities]

The library is essentially finished.

It has passed all tests on about 20 different
operating system / compiler / processor combinations
(thanks to the patient help of Boris Gubenko, John Maddock, Markus Schöpflin and others).

The only missing feature, that I can think of,
is a mechanism for having the library use
the isnan, signbit etc provided by the compiler
instead of its own isnan, signbit etc implementations.
(The default might be that if a compiler is C99 compatible,
then the functions provided by the compiler are used,
otherwise the libraries own implementation is used.)

There is also an overlap with the Math Tool Kit library;
both libraries provide implementations of fpclassify, isnan etc.
(The two implementations use completely different approaches.)

My plan has been to have the library included in Boost as an addition to Boost.Math
rather than as an independent library.

However, I'm very busy; the next year will probably decide whether my company
becomes a success or a failure.

Is anyone interested in taking over the library?
If needed, I would be happy to help that person get started.

--Johan

_______________________________________________
(Continue reading)

Permalink | Reply | 2 comments |

Gmane